Test for additive interaction in proportional hazard model applied to Pitot sensors reliability and survivability

Cox proportional hazards model is one of the most common methods used in analysis of time to event data. Estimation of proportional hazards models when covariates are time independent increase the power of significance of the model. Techniques to release this assumption of proportionality allow researchers to test whether the effects of covariates change over time and also permit a more nuanced understanding of the phenomenon being studied. The current paper presents an extension for the Cox model in case of non-proportional risk. This new model will be applied in the Pitot Sensor's reliability function to study the covariates interaction with time. A simulation study is used to illustrate the model and to show its new performances. This paper focuses in the determination of the Pitot sensor reliability and survivability Function. However, the same process can be generalized to study the reliability function of other standby systems.